Metro taller transmission swap thread: +5% efficiency gain (taller final drive)
 

(Originally written: 05-29-2006)

---

Motivated by SVOBoy's transmission swap...

Exactly when I would make the time to do this, I don't know. I already have too many projects stacked up in a holding pattern over the airport.

But, I figure: start this thread and maybe it'll motivate me or someone else.


There it is - transmission from a 4-cylinder (non-GT) Suzukiclone. Same housing and gear ratios as the 3-cylinder version, just a taller final drive:

1998 Metro/Firefly 1.0 L:
1st = 3.416
2nd = 1.894
3rd = 1.280
4th = 0.914
5th = 0.757
Final = 4.39

1989-1994 Swift 1.3L SOHC 5-speed:
1st = 3.416
2nd = 1.894
3rd = 1.280
4th = 0.914
5th = 0.757
final = 3.523

gear ... mph ... ratio ... rpm @ 4.39 calculated (observed) ... rpm @ 3.523

2 ... 40.4 ... 1.894 ... 4959 (5175) ... 4157
3 ... 40.4 ... 1.280 ... 3351 (3495) ... 2809
4 ... 40.4 ... 0.914 ... 2393 (2500) ... 2006
5 ... 40.4 ... 0.757 ... 1982 (2080) ... 1591

Engine RPM figures are calculated from this site:
http://wildcatent.freeyellow.com/zookmods/calc.htm
using outside tire diameter of 22.764 inches, calculated here:
http://www.net-comber.com/tirecalc.html

(Observed RPM) may not be reliable because it's hard to get a figure off the ScanGauge - digital tachs suck, especially when the numbers move around.

Works out to a caculated 391 rpm drop going from the stock 4.39 final drive to 3.523 in 5th gear at 40.4 mph

There was an observed 420 rpm drop on the stock 4.39 when shifting from 4th to 5th @ 40.4 mph.

The mini experiment showed a 8.9 mpg/18.9% improvement in FE from that 420 rpm drop at that speed.

Something to think about anyway.

Some more to think about: this car's 0-60 time in stock form is about 15 seconds. This swap will make it ... much worse! One teamswift member found this swap intolerable.

Then there's the issue of the speedometer/odometer being off. Can compensate for that in the ScanGauge though.

It's all about the torque. I guess that's possible. But where to get the gears? Also, that would mean taking the tranny apart, which is slightly scarier and requires special tools, I think.

A straight swap should theoretically only take a couple of hours:

- remove starter
- detatch steel lower rad tube bolted on top
- disconnect clutch cable, reverse switch wires, speedo cable
- remove axles
- support engine
- unbolt & remove tranny

Sounds like the "nerd gear" option that my brother in law and I talked about. As in - why wouldn't the manufacturers offer a taller that stock top gear for the hard-core FE nuts?

Yes it would force you to shift more (in and out of top gear) depending on terrain, speed, cargo/passenger load.

But if you're willing to tick the nerd gear box on the option sheet, I'd bet you'd also be willing to shift more for the FE advantage if it's right for your situation.

Lighter, less rotating mass (12 in wheels & tires); also the pre-86 cars EPA ratings were higher, even though they used the same test method (they introduced a "fudge factor" and changed the ratings even though the cars didn't change).

As for the gear ratios, everything I've learned about those trannies, I learned at TeamSwift. If I recall rightly, all the 3-cyl cars had the same ratios, but used different final drives depending on: XFi/12 in rims/13 in rims/convertable (heavier).

Some bedtime reading...

http://www.teamswift.net/viewtopic.php?t=14845
http://www.teamswift.net/viewtopic.php?p=163307
http://www.teamswift.net/viewtopic.php?t=23748

Increasing wheel/tire size would have the same effect. And truth be told, if I knew someone with a big honking set of 14 or 15 inch wheels that would fit our odd-ball bolt pattern, I would like to try them first and do some calculations to see whether or not I'd be happy with the transmission swap.

I probably wouldn't be doing any of this if I didn't have the transmission. It was essentially free.

Well, there's no doubting it would help my overall mileage. The only experimental question is whether it would hurt the car's driveablility so much that I wouldn't like it. That's why I would just try the oversized wheels/tires if I had access to some that would fit.

Assuming I was OK with the driveability, I would still switch back to the smaller wheels and tires and do the swap because of the aero, LRR, and rotating mass advantages of the OEM tires/wheels setup (assuming the bigger wheels/tires combo is heavier).

Rotational (ie. drive train) mass takes much more energy to accelerate and brake than other vehicle mass - 7 times as much, if you believe this page:
http://www.my330i.com/mod19.php - So reducing rotational mass will particularly help if your driving is not primarily steady speed.

Weak synchros are the biggest problem with these cars. I've driven 7 different versions of these cars, and four of them had crunchy 2nd gears or balky 3rd gears. Even in my 98, which is essentially "brand new" (just passed 8000 km yesterday), I have to be careful changing up to 2nd: I can crunch it if I switch too quickly (ie. "normal" speed in any other car), especially when cold.
You are correct, sir. I don't care about speed or acceleration. If I did, I never would have bought the Fireflea in the first place.

However, one concern of going with a too-high final drive is clutch wear. Starting from a stop will require more slip. One of the TeamSwift posters said starting on a hill with the 13 in tires and diff from the 4-cyl car would not be good news at all. We'll see. Perhaps.

Some more thinking out loud... (have a highway trip coming up in the next couple of weeks, which gets me thinking about aerodynamics and gear ratios)

If I had a hypothetical 1st gear ratio of 3.0 and a final drive of 4.0, that means my engine turns 12 times (3.0 x 4.0) for each turn of the axle, correct?

If that's the case, then the taller diff on the 4-cyl tranny isn't horribly different than the 3-cyl car. From a few comments I've read, it sounded like starting out in first gear would be made much harder - like the equivalent of starting in second gear all the time with the stock tranny.

But if I compare the 1st and 2nd gears of the stock tranny to the 1st gear of the taller one, I get:

3-cyl tranny: gear 1 (3.416) x final drive (4.39) = 15.0 wheel revolutions per engine revolution

3-cyl tranny: gear 2 (1.894) x final drive (4.39) = 8.3 wheel revolutions per engine revolution

4-cyl tranny: gear 1 (3.416) x final drive (3.52) = 12.0 wheel revolutions per engine revolution

By that math, it's not quite like starting in my current 2nd gear. It's roughly in the middle of my existing gears 1 & 2 (slightly closer to 1).

Where ERPWR = "engine revs per wheel rev" (gear ratio * final drive ratio)

gear ... gear ratio ... ERPWR, stock tranny ... ERPWR, 3.52 final (4 cyl) tranny

1 ... 3.416 ... 14.99624 ... 12.02432
2 ... 1.894 ... 8.31466 ... 6.66688
3 ... 1.28 ... 5.6192 ... 4.5056
4 ... 0.914 ... 4.01246 ... 3.21728
5 ... 0.757 ... 3.32323 ... 2.66464

So I'm kind of "gaining" another gear. The stock tranny 5th gear/diff product is actually slightly shorter than the product of the 4-cyl tranny's 4th gear/diff.